Operational reliability evaluation and analysis framework of civil aircraft complex system based on intelligent extremum machine learning model

To reasonably implement the operational reliability analysis and describe the importance of the influencing parameters for the operation status, a framework for operational reliability evaluation and analysis is proposed. First, an operational reliability evaluation model (OREM) is established based on the data envelopment analysis (DEA) method, which takes quick access recorder (QAR)data as input to comprehensively evaluate the operational characteristics of complex systems to obtain the operational reliability P_r. Then, to enhance the modeling efficiency and simulation performance for the operational reliability analysis of the complex system, we propose an intelligent extremum machine learning model (IEMLM), by integrating extremum response surface method (ERSM), artificial neural network (ANN), improved particle swarm optimization (PSO) algorithm, and Bayesian regularization (BR) algorithm. The operational reliability analysis of a braking system of a civil aircraft is conducted to validate the effectiveness and feasibility of this developed method, by considering the comprehensive influence of system-environment-human. The comparison of IEMLM, RF, and ANN shows that IEMLM improves the analysis accuracy and calculation efficiency. The proposed framework and models can provide useful references for civil aircraft operational reliability analysis, special situation treatment, maintenance, and design.